[02:39] roaksoax: fwiw I am getting apparmor problems with dhcp on quantal using the dailybuild package, is this known? [02:40] bigjools: no tomy knowledge. what kind of error? [02:41] roaksoax: Apr 9 12:35:44 maas dhcpd: Open a socket for LPF: Permission denied [02:41] Apr 9 12:35:44 maas kernel: [ 1194.554101] type=1400 audit(1365474944.338:53): appar [02:41] mor="DENIED" operation="create" parent=1 profile="/usr/sbin/dhcpd" pid=3444 comm="dhc [02:41] pd" family="packet" sock_type="raw" protocol=768 [02:42] bigjools: has packaging changed in upstream? cause trunk should have all of the fixes that were made for qusntal [02:42] roaksoax: dunno - this is a build from a month ago [02:42] roaksoax: the daily ppa is a mess of versions [02:43] could do with an overhaul [02:43] bigjools: thats weird [02:43] yeah [02:43] anyway, testing the experimental Go-juju provider [02:43] bigjools: i have not sern those issues running raring nor the sru prepared package [02:45] roaksoax: I suspect the package itself is screwed, I don't trust them any more because of so many version problems (upload fail with duped files) [02:45] this makes me sad [02:45] yeah [04:51] roaksoax: commissioning and enlistment are taking *ages* now... :/ [07:32] i can't get the MAAS notice "The region controller does not know whether any boot images have been imported yet. If this message does not disappear in 5 minutes..." to go away [07:32] where are these logs I'm supposed to inspect? [07:53] anybody here? [13:26] bigjools: i think I know why that is.. and I already have a fix for it [13:46] allenap: howdy!! So FPI is mostly done, I do, however, have a question for you [13:46] allenap: check https://code.launchpad.net/~andreserl/maas/trunk-fpi/+merge/152039 and look at line 707 [13:46] allenap: how do you think we should approach obtaining the interface in order to obtain the IP when there are no managed networks? === matsubara_ is now known as matsubara [14:05] roaksoax: I can't think of a better way than that. Do you have anything in mind? [14:08] allenap: no idea.. wsa trying to think of a way but other than obtaining the first one, there doesn'y seem to be much of a solution [14:10] roaksoax: Stick with that for now then. [14:16] allenap: will do, thanks! [17:12] hello! [22:50] roaksoax: would be be in the ipmi detection by any chance? [22:50] s/be be/it be/ [22:52] bigjools: were you using a branch with the new ipmi detection? [22:53] bigjools: or did you enable the IPMI params? [23:10] roaksoax: argh, sorry, desktop problems. my mouse keeps disconnecting itself and replugging doesn't work, I need to reboot. [23:10] anyway, it is slow with or without ipmi params enabled [23:12] bigjools: have you grabbed any logs? (such as cloud init ones?) [23:13] bigjools: while i have runned into slow ipmi detection myself, Its for network issues as my network setup is a strange one [23:14] roaksoax: not yet, fighting maas [23:14] and I need the backdoor image... grar [23:21] bigjools: re on your re-writing thing.. I agree. I wanted to separate the ipmi detection on a separate package/binary... however, the particular bug that's been fixed there is a bug that was targetted for 1.2 and has been there for a long time [23:21] and the only thing that caused is longer time on enlistment/commissioning for trying to discover ipmi [23:25] roaksoax: only longer on VMs though? [23:25] it takes about 10 minutes to complete on the HP microservers now [23:26] bigjools: with today's fix, it should not even try to detect ipmi in VM's [23:27] bigjools: and that's weird, I have my microservers here working just fine [23:27] bigjools: they were taking longer than usual only because of my network setup [23:27] roaksoax: I suspect my maas is screwed up, I had to downgrade because of the crazy package versioning and it may have gone wrong [23:27] I will reinstall [23:28] bigjools: ok [23:28] bigjools: what are you using, daily builds? [23:28] yeah [23:28] weird then [23:28] I previously had a home-made trunk build on quantal [23:28] I switched to daily [23:29] but the south migrations in the DB was screwed [23:29] yeah that's a likely cause of nighmare's [23:29] * bigjools getting fed up of twitter follow spam [23:47] bigjools: i'll be on and off for the next hour/2 hour so let me know your findings [23:47] roaksoax: I'll be a while - I am upgrading the server to raring